My understanding is that L's and W's are both bad because they lead to a timing flaw - namely that the pitcher's torso starts to rotate forward while the arm is still moving up and back.
Just copy a couple of Chris O'Leary pictures, an inverted W like this:
View attachment 79388
leads to a situation like this:
View attachment 79389
where the the foot has hit the ground and the torso is starting to rotate forward while the arm is still moving up and back in order to get into cocked position. When the arm finally reaches cocked position:
View attachment 79394
the torso is already moving forward (you can see the belt buckle is already facing the hitter), which means that the arm now has to change directions and whip forward...causing a ton of stress on the elbow.
Meanwhile, someone like Justin Verlander already has his arm up as his foot is striking the ground:
View attachment 79392
which means no major changes of direction for his arm/elbow.
Anyway, as far as the harder offspeed stuff, one thing I've been wondering about is whether the grips/different approaches are causing issues with respect to pronation - i.e. the idea that once you let go of the ball, your wrist should naturally rotate in towards your body, as shown here:
View: https://www.youtube.com/watch?v=iIenH2Kt0rg
Pronating lets your forearm take some of the load off your elbow when you decelerate your arm. Note that in the video, the pitchers (Kershaw and the pitching coach) are throwing curveballs, which spin in the opposite direction from the desired pronation (i.e. they spin away from the body), but the pitchers are still pronating after release. Fastballs are easier to pronate on because the rotation of a fastball just follows your natural arm action.
One potential problem is that if you start out with your wrist already rotated, you can't pronate when you're decelerating (your wrist has nowhere to go), which means that the elbow is taking the brunt of the deceleration. A telltatle sign for this is if the pitcher is pointing the ball to second base in the delivery; if they're doing that, they've already rotated their wrist and can't pronate much more. Here are Ivan Nova and Mariano Rivera showing "point the ball to second" vs. the safer "point the ball to third" techniques:
View attachment 79343
Similarly, if we compare Jacob DeGrom with Mariano Rivera at similar points (just before foot strike) in their delivery:
View attachment 79342
View attachment 79344
As mentioned above, breaking stuff also requires more care to be taken to ensure pronation, since the ball has to leave the hand before the arm can pronate (otherwise you're putting the wrong spin on the ball.) Which leads me to an
NIH study I was reading:
My horribly uninformed guess is that harder breaking stuff and/or different grips are causing pitchers to pronate less effectively, either by changing the way the ball is held or increasing the amount of time spent in supination, thereby making elbows more likely to blow up.
[Edited to add pictures that actually load]